An AI tool that can predict the risk of a heart attack over a 10-year period could change the way patients undergo CT scans to check for chest pain, according to new research funded by the company. The study was presented today at the American Heart Association’s Scientific Sessions in Philadelphia.

The first real-world trial of an AI tool found it could improve treatment for up to 45 per cent of patients and potentially save lives for people living with chest pain. heart attack.

Around 350,000 people in the UK undergo a cardiac CT scan each year. This is a standard test to identify narrowings or blockages in the coronary arteries. In about three-quarters of cases, there are no obvious signs of significant stenosis, so patients can often be discharged home with peace of mind. Unfortunately, many of these people will have a heart attack in the future. This is because when it becomes inflamed and clogs the artery, tiny invisible narrowings can break open. Until recently, it was not possible to identify patients at risk.

“Utilize” artificial intelligence

The new study, led by Professor Charambos Antoniades from the University of Oxford, analyzed data from more than 40,000 people who regularly undergo cardiac CT scans at eight UK hospitals. Participants were followed for a median of 2.7 years. They found that while patients with significant narrowing of their coronary arteries were more likely to have a serious cardiac event or death, they were twice as likely as patients without significant narrowing to experience a heart attack or death. discovered.

The team used a new AI tool that was trained with information about fatty changes around inflamed arteries, as well as information about artery narrowing and other clinical risk factors. Further long-term testing revealed that it could independently and accurately predict the risk of cardiac events.

Among people without blockages in their arteries, those with the highest levels of inflammation in their blood vessels had a more than 10 times higher risk of cardiac death than those with lower levels of inflammation.

world’s first pilot

In a world-first pilot, the team provided clinicians with AI-generated risk scores for 744 patients. We found that clinicians changed the patient’s treatment plan in up to 45% of cases, demonstrating that this AI tool could be of great value in the following areas: Provide guidance and information on how to manage patients with chest pain, ensuring early identification and preventive treatment of patients most at risk.

Analysis comparing the use of AI tools to standard care reveals that they are highly cost-effective for the NHS. Furthermore, the researchers estimate that introducing the technology to the NHS could reduce heart attacks by more than 20 per cent and heart deaths and strokes by 8 per cent in people tested. The technology needed to power this AI tool has already been commissioned by NHS England for a pilot program in five NHS hospitals, and researchers believe the tool could soon be rolled out across the UK. I hope there is.

Professor Charalambos Antoniades, BHF Professor of Cardiovascular Medicine at the University of Oxford, said: For the next 10 years, even if there are no signs of disease in the arteries of the heart. Here we demonstrate that providing clinicians with an accurate picture of risk has the potential to change and improve treatment for many heart disease patients.

“We hope this AI tool will soon be rolled out across the NHS and help prevent thousands of avoidable heart attack deaths in the UK every year.”

Professor Sir Nilesh Samani, our Medical Director, said: “This study shows that AI-based technology can help better identify patients at highest risk of future heart attacks, thereby helping clinicians make better treatment decisions for them. It shows that they can play a valuable role.”

“Too many people die needlessly from heart attacks every year. We need to harness the potential of AI to guide patient care and ensure that the NHS is equipped to support its use. It is important to ensure that this technology is ultimately rolled out across the NHS, saving thousands of lives every year that could otherwise go untreated. I hope it helps.”

Professor Antoniades’ research is also supported by the National Institute for Health and Care Excellence’s Oxford Biomedical Research Center.

Learn more about our research